Mario Antonini Hemodynamic and metabolic effects of transjugular intrahepatic portosystemic shunt (TIPS) during anesthesia for orthotopic liver transplantation

Giorgio Della Rocca Francesco Pugliese Livia Pompei Micaela Maritti Cecilia Coccia Alessandro Gasparetto Raffaello Cortesini

Received: 4 July 1995 Received after revision: 16 November 1995 Accepted: 20 December 1995

M. Antonini (m) . G. Della Rocca F. Pugliese . L. Pompei M. Maritti . C. Coccia A. Gasparetto University of Rome “La Sapienza”, Institute of Anesthesiology and Reanimation, Viale del Policlinico, 1-00 161 Rome, Italy Fax: + 39 6 446 3661

R. Cortesini University of Rome “La Sapienza”, I1 Department of Surgery, Wale del Policlinico, 1-00 161 Rome, Italy

Abstract Recently, the transjugu- lar intrahepatic portosystemic shunt (TIPS) has been advocated as a safe bridge to orthotopic liver transplan- tation (OLT). We retrospectively studied 53 consecutive cirrhotic pa- tients who underwent OLT 27 pa- tients with TIPS were compared to 26 controls. Hemodynamic and oxy- phoretic data (Fick method) were collected during six phases of OLT. There were no significant differ- ences in demographic data and Child-Pugh class, nor in surgical time and blood product require- ments before the anhepatic phase between TIPS patients and controls. In the TIPS group, we observed a marked hyperdynamic profile with a lower systemic vascular resistance index, higher cardiac index, and de-

pressed oxygen consumption before native liver removal. During the same period, the TIPS group devel- oped a greater acidosis and was treated with a larger amount of Na- HCO,. Following the anhepatic phase, no differences between the two groups were detected. All transplantations were successful, and no complications related to TIPS were observed. These results seem to be the consequence of a re- duced liver function reserve with a direct hemodynamic effect due to the TIPS.

Key words TIPS, liver transplantation . Liver transplantation, TIPS . Portosystemic shunt, liver transplantation

Introduction

The transjugular intrahepatic portosystemic shunt (TIPS) is a reliable bridge to liver transplantation in pa- tients with variceal bleeding. The TIPS is an expand- able, flexible, metallic stent placed between a branch of the portal vein and the systemic venous system, and it reduces portal pressure and the risk of bleeding. TIPS therapy is highly successful and has a low incidence of complications. Since 1992 the number of liver transplan- tations in patients with TIPS has increased at our insti- tution. The aim of the present study was to analyze the impact of TIPS on heemodynamics, oxyphoresis, and acid-base status during liver transplantation.

Methods

The records of 53 consecutive cirrhotic patients who underwent or- thotopic liver transplantation (OLT) with the same surgical and anesthesiological staff at University of Rome “La Sapienza” be- tween January 1991 and April 1995 were retrospectively studied. This research was approved by the local Ethics Committee, and in- formed consent was obtained from all patients. The study included 27 TIPS patients (group TIPS) and 26 control patients (group C). Demographic data, etiology of the liver disease, Child-Pugh classi- fication, and relevant clinical findings are shown in Table 1. The temporal distribution of OLT in the two groups is shown in Fig. 1. The indications for TIPS were not selective. TIPS surgery was per- formed with Wall ( n = 17), Nitinol-Strecker ( n = 9), and Palmaz ( n = 1) stents; six patients required balloon dilatation or an additio- nal stent implantation. All patients showed an immediate reduc- tion in portocaval pressure gradient greater than 50 YO; moreover, these patients were completely free of hemorrhage while ascites

404

Table 1 Demographic data, Groups TIPS C etiology, Child-Pugh classifica- tion, and other diagnostic data Patients Number 27 26 in TIPS and control (C) groups Sex 23 M-4 f 20 M-6 F (PHC, posthepatitis cirrhosis; Age in years (range) 48 (32-62) 42 (28-62) PHC-K, carcinoma in posthe- Weight in kg (SD) 72 (11) 73 (12) patitis cirrhosis) Etiology PHC

PHC-K Other

C Child-Pugh classification B

21 5 1

20 7

23 2 1

21 5

a Some patients have multiple Diagnostic datab Variceal bleeding 14 (52 Yo) 1.5 (58 Yo)

Renal failureb 8 (30 Yo) 9 (35 %) diagnoses Ascites 17 (63 %) 15 (58 Yo)

Creatinine clearance < SO ml . min-l

--r---, 1991 1992 .__

(Apr.1 Fig.l Temporal distribution of liver transplantation in TIPS and control ( C ) groups

tablished throughout the entire procedure. Inotropes were never used. Venovenous bypass (Biomedicus) was done in all cases.

Hemodynamic and oxyphoretic data with acid-base status (pH, BE) and NaHCO, administration were collected during six differ- ent phases: basal (after anesthesia induction and before isoflurane administration), hepatectomy, preanhepatic (at onset of veno- venous bypass), anhepatic, early neohepatic ( 5 min after graft re- vascularization), and neohepatic. Moreover, surgical hepatectomy time and blood product requirements were determined before the native liver was removed. All values are shown as mean (SD). Car- diac output was measured in triplicate by thermodilution. The fol- lowing parameters were calculated according to body surface area: cardiac index (CI), systemic (SVRI) and pulmonary (PVRI) vascular resistances, arterial oxygen delivery (O,AVI), and oxygen consumption (V0,I). These parameters and the oxygen extraction ratio (0,ER) were obtained by standard computerized formulas (software CLICS Spacelabs). Statistical significance was assessed using Student's t-test for unpaired data (SPSS plus). A P value less than 0.05 was deemed significant.

and renal function always improved. The median time interval be- tween the TIPS procedure and liver transplantation was SO days (range 1-270 days). TIPS patency was assessed by Doppler color ultrasound when patients were scheduled for OLT and confirmed by examination of the hepatectomy specimen.

All patients received oral benzodiazepines as premedication, and general anesthesia was induced with fentanyl(5 pg . kg-'), dro- peridol (0.04 mg . kg-I), and sodium thiopental (2-3 mg . kg-I). Pancuronium bromide (0.08 mg . kg-') was used as a muscle relax- ant. Anesthesia was maintained with isoflurane (0.8 %-1.2 YO) sup- plemented by fentanyl in 0, air (FiO, = 0.5). Mechanical ventila- tion, with positive end-expiratory pressure equal to 5 cmH,O, was performed with a Siemens Servo 900D to obtain mild hypocapnia (ETCO, 4.8-5 kPa). To maintain body temperature, all patients were placed on a heating blanket and all of the intravenous lines were warmed. Lost blood was replaced with packed red cells (RBC) to maintain the hemoglobin level at 10 g . dl-'. Volemic re- placement was obtained with normal saline and fresh frozen plas- ma (FFP) in accordance with filling pressures and laboratory data. Calcium chloride was administered in boluses as indicated by the Ca + + plasma levels. Base excess (BE) values determined sodium bicarbonate administration. A continous infusion of dopamine (2 pg. kg-' . h-') and aprotinin (500000 u . h-I) was es-

Results

All patients in both groups had successful transplanta- tions, and complications due to the TIPS were not de- tected. The surgical hepatectomy time was 188 min (SD 39) in the TIPS group and 212 min (SD 49) in the con- trol group ( P = NS). The total blood product require- ments until the anhepatic phase were, respectively, 2.6 1 (SD 1.7) in the TIPS group and 2.5 1 (SD 1.7) in the C group, without statistical differences. Hemodynamic and oxyphoretic data are shown in Tables 2 and 3. In the basal, hepatectomy, and preanhepatic phases, a sig- nificantly lower SVRI was observed in the TIPS group than in the control group. In TIPS patients CI, mean pulmonary arterial pressure (MPAP), and (PAWP) were significantly higher during the hepatectomy phase. V0,I was lower in the basal and hepatectomy phases in the TIPS group, although 0,ER was lower in the hepa- tectomy phase only. As for the acid-base status (Fig. 2), base excess and pH were significantly lower in the TIPS group during the preanhepatic phase ( P < 0.01).

405

Table 2 Hemodynamic parameters In the different phases of OLT

SVRI, systemic vascular resistance index; CVP, central venous

Phases Group CI

pressure; MPAP, mean pulmonary arterial pressure; PVRZ, pul-

HR PAWP MAP SVRI CVP MPAP PVRI

(Cf,cardiac index; H R , heart rate, J MAP, mean arterial pressure; monary vascular resistance index)

(1 . min-' . m-2) (beats. min-') (mmHg) (mmHg) (dyne. sec . (mmHg) (mmHg) (dyne. sec . cm-' . m-*\ cm' . m-*\

Basal TIPS 5.2 (1.5) C 4.9 (1.8)

Hepatectomy TIPS 6.3 (1.2)* C 5.5 (1.4)

Preanhepatic TIPS 6.0 (1.4) C 5.4 (1.3)

Anhepatic TIPS 4.8 (1.1)

Early TIPS 7.1 (1.6) neohepatic C 7.1 (1.6) Neohepatic TIPS 6.3 (1.3)

C 5.8 (1.3)

C 4.3 (0.9)

89 (15) 95 (21) 99 (10) 93 (13)

103 (11) 94 (13)

104 (10) 97 (15)

103 (9) 100 (14)

97 (14) 99 (9)

11 (3) 76 (16) 1152 (337)' 6 (3) 16 ( 5 ) lO(4) 85 (23) 1394(421) 6(2) 16(5) 13 (3)' 81 (9) 978 (251)* 8 (2) 18 (3)* 11 (4) 85(12) 1207(391) 7(3) 16(4) 10 (3) 79 (10) 1013 (290)* 8 (3) 14 (4) 9(4) 84(11) 1219 (411) 6(4) 13 ( 5 )

7(2) 82(10) 1492(437) 6 (2) 12(3) 14 (6) 78(11) 809(242) lO(4) 20(7) 13 (5 ) 78(14) 819(253) lO(4) 19 (6) 13 (3) 81 (10) 962 (233) 8 (2)" 19 (4) 11 (4) 79(13) 1048(347) 6(3) 17(4)

7(3) 82(11) 1327 (360) 8 (3) 12 (3)

75 (37) 101 (67) 70 (23) 83 (37) 66 (31)

88 (35) 101 (42) 74 (28) 78 (39) 74 (27)

93 (51)

79 (35) * P < 0.05; ** P < 0.01

Table 3 Mixed venous satura- tion ( S V 0 2 ) , arterial oxygen delivery (0, AVI), oxygen consumption (VO, I), and oxy- gen extraction ratio (0, ER; Fick method) in the different phases of OLT

* P < 0.05; ** P < 0.01

Phases Group SVO, 02AVI VOJ 0,ER

Basal TIPS 88.1 (4.6) 747 (245) 98 (19)** 14.1 (4.0) C 85.9 (4.9) 839 (300) 122 (31) 15.8 (4.8)

("/I (mi . min-' m-2) (ml . min-' . m?) (yo)

Hepatectomy TIPS 90.3 (3.6) 895 (216) c 88.1 (3.9) 920 (241)

Preanhepatic TIPS 90.6 (3.0) 790 (190) C 88.8 (4.8) 824 (218)

Anhepatic TIPS 90.4 (2.9) 652 (137) C 89.2 (3.3) 658 (136)

Early TIPS 89.1 (6.1) 995 (242) neohepatic C 89.3 (3.7) 1088 (215) Neohepatic TIPS 88.4 (4.0) 851 (176)

C 87.1 (4.0) 841 (1781

Sodium bicarbonate administration (mEq) in the TIPS and control groups was, respectively, 54 (SD 60) vs 20 (SD 49) during the hepatectomy phase ( P = 0.028) and 53 (SD 54) vs 10 (SD 28) during the preanhepatic phase ( P < 0.001; Fig. 2). No differences between the two groups were detected in the anhepatic phase or after li- ver graft reperfusion.

Discussion

The role of TIPS as a bridge to transplantation, if con- firmed in the future, will increase the number of OLTs in patients with intrahepatic portosystemic shunts be- cause these stents remain functional longer than 4 years [2]. TIPS reduces portal venous pressure and blood flow through esophagogastric varices, reducing the risk of bleeding. TIPS is of particular value in the treatment of

105 (19)" 122 (23) 104 (28) 113 (36) 90 (26) 90 (1 9)

139 (39) 147 (32) 129 (31) 132 (26)

12.4 (3.2)' 14.5 (4.3) 13.4 (2.8) 14.4 (4.6) 13.6 (2.6) 14.1 (3.2) 14.7 (5.6) 13.9 (4.1) 15.6 (3.8) 16.2 (3.6)

clinical ascites; good results have also been reported in patients with both ascites and hepatorenal syndrome [6], as our experience also confirmed. TIPS is a much less invasive procedure than surgical portosystemic shunt implantation and, because the totally intrahepatic stent is removed with the native liver, there is no further technical complication during liver transplantation [4, 81. Although Martin et al. [6] and Woodle et al. [9] re- ported that TIPS markedly reduced surgical operative time and transfusion requirements during OLT, in agreement with Freeman et al. [3] we did not observe any significant differences in either case. On the other hand, deterioration of hepatic function as a result of de- creased portal venous perfusion is a potential risk. How- ever, hepatic synthetic function is generally maintained because TIPS improves nutritional status reducing por- tal hypertension and mobilizing ascites [2]. Martin et al. [6] noted a progressive loss of liver volume, a pro-

406

-10 1

* 120 i - ).

0"

z mLhLh 40 0

&t.l mpodmy Pranheprm

!,,, 7.43 (0.07) 7.40 (0.06) 7.37 (0.05 7.41 (0.06) 7.41 (0.04) 7.41 (0.04

0.1 (4.0) -2.8 (3.0) -5.2 (2.6Y

-1.4 (2.7) -2.1 (2.4) -2.6 (2.0)

pH

T !

I i

LnhsPic

,40 (0 05,

38 (0.051

~

~

.

-4.0 ( 2 3

-4.4 (2.4)

67 V3)

~

~

~

84 (84) ~

N&*@,C

7.40 (0 06)

-3.5 (3.1)

-3.7 (2.8)

Fig.2 Base excess and NaHCO, administration during liver trans- plantation in TIPS and control (C) groups. pH is shown in the table at the bottom

longation of caffeine-antipyrine clearance, and an in- creased incidence of encephalopathy without any signif- icant change in serum bilirubin, prothrombin time, or al- bumin levels. Although TIPS is a calibrated portosys- temic shunt, it presents the disadvantages of the non-se- lective shunts, i.e., a gradual decrease in hepatic mass and quantitative hepatic function [6]. Therefore, the in- dication in patients with a poor hepatic functional re- serve is highly questionable [l] and may be justified only in emergencies.

The aim of our study was to retrospectively compare intraoperative hemodynamic effects of TIPS during OLT between two groups of cirrhotic patients treated during different periods of time, as we have been experi- encing a progressive replacement of cirrhotic patients with TIPS-treated cirrhotic patients as transplant candi- dates. Nevertheless, we believe that our observations

are of some value because: a) both study groups showed comparable severity in liver disease according to their Child-Pugh classification, (b) surgical and anesthesio- logic management were identical, (c) hepatectomy time and blood requirements did not show significant differ- ences, and (d) all differences between groups were elim- inated with removal of the native liver.

Ozier et al. [7] demonstrated in anesthetized patients that TIPS placement induced an increase in preload, CI, and 0,AVI without changes in VOJ, measured with in- direct calorimetry. Only a few of these observations were confirmed in the preoperative hemodynamic pro- file of liver transplant patients studied by Lopez-Ol- aondo et al. [5], who did not find any differences in CI or peripheral resistances between TIPS patients and controls. Our results suggest a specific intraoperative hemodynamic pattern in patients with TIPS before the anhepatic phase showing a marked hyperdynamic pro- file with a lower SVRI, higher CI, and reduced V0,I. The significantly higher values of PAWP and MPAP during hepatectomy could have been a direct effect of the intrahepatic stent. As the native liver was removed, the two groups became absolutely identical. The differ- ences in hemodynamic and oxygenation data could only have been due to a higher degree of liver failure in TIPS patients. Nevertheless, TIPS may affect hemo- dynamic and oxyphoretic parameters in such critical sit- uations as isoflurane anesthesia, positive pressure venti- lation, and surgery. As a clinical consequence, a larger amount of sodium bicarbonate was administered in the TIPS group before the anhepatic phase, even though it was not sufficient for a complete correction of acidosis at the onset of venovenous bypass (preanhepatic phase).

In conclusion, TIPS is effective in controlling ascites and bleeding varices in patients with acceptable liver function who are awaiting transplantation. During liver transplantation patients with TIPS become more aci- dotic before the anhepatic phase and have a marked hy- perdynamic circulation as well as a reduced oxygen con- sumption. The lack of intraoperative surgical complica- tions related to TIPS that was observed in our series gives TIPS a great advantage compared with classic portocaval shunts. The expected decrease in surgical op- erative time and transfusion requirements have yet to be confirmed.

References 1. Blei AT (1994) Hepatic encephalopa-

thy in the age of TIPS. Hepatology. 20: 249-251

2. Conn H O (1993) Transjugular intrahe- patic portal-systemic shunts: the state of the art. Hepatology 17: 148-158

3. Freeman RB, Fitzmaurice SE, Green- field AE, Halin N, Haug CE, Rohrer RJ (1994) Is the transjugular intrahepatic portocaval shunt procedure beneficial for liver transplant recipients? Trans- plantation 58: 297-300

4. Hoek B van, Lindor KD, Murtaugh PA, Harrison JM, Krom RAF, Wiesner RH, Nagorney DM (1994) The role of por- tosystemic shunts for variceal bleeding in the liver transplantation era. Arch Surg 129: 683-688

407

5. Lopez-Olaondo L, Monedero P, Rios J de los, Iribarren MJ, Hidalgo F, Skez A (1994) Anaesthetic implications of transjugular intrahepatic portosystemic shunt in orthotopic liver transplanta- tion. Br J Anaest 72s: A116

6. Martin M, Zajiko AB, Orons PD, Dodd G, Wright H, Colangelo J, Tartar R (1993) Transjugular intrahepatic porto- systemic shunt in the management of variceal bleeding: indications and clini- cal results. Surgery 114: 719-727

7. Ozier Y, Moutafis M, Legmann P, Mar- tin-Bouyer Y, Conseiller Ch (1994) Tis- sue oxygenation in anesthetized cir- rhotic patients during transjugular in- trahepatic portosystemic stent-shunt (TIPSS) procedure. Anesthesiology 81: A283 344-351

8. Ring EJ, Lake JR, Roberts JP, Gordon RL, La Berge JM, Read AE, Sterneck MR, Ascher NL (1992) Using transjug- ular intrahepatic portosystemic shunts to control variceal bleeding before liver transplantation. Ann Intern Med 116:

9. Woodle ES, Darcy M, White HM, Per- driset GA, Vesely TM, Picus D, Hicks M, So SKS, Jendrisak MD, McCullogh CS, Marsh JW (1993) Intrahepatic por- tosystemic vascular stents: a bridge to hepatic transplantation. Surgery 113:

304-309